Holoprosencephaly represents a common birth defect (1:16,000 in live births and 1:250 in stillbirths) with a broad spectrum of craniofacial malformations ranging from distressful cyclopia to mild symptom of a single central incisor. It is caused by defects in the specification of the ventral forebrain (a part of the anterior axial midline), which subsequently lead to incomplete separation of the brain into the left and right hemispheres. Recent studies indicated that Nodal signaling plays a central role in controlling midline development, we therefore will focus on the regulation of Nodal signaling in mouse embryogenesis with a special interest in anterior axial midline formation. Nodal is a member of the transforming growth factor beta (TGF-beta) superfamily that utilizes a signaling pathway defined by Activin type I and II receptors, Smad2 and 4, and FoxH1(FAST). Importantly, members of the EGF-CFC family of extracellular proteins such as mouse Cripto are essential co-factors for Nodal. We previously reported a Cripto null allele, and recently we generated a Cripto hypomorphic allele, Cripto3-loxP, by genetic manipulation. Approximately 50% of the Cripto3-loxP/CriptoNull mice displayed a wide range of axial midline defects resembling holoprosencephaly. In contrast, TGIF is a homeobox gene encoding a nuclear protein that antagonizes TGF-beta signaling by blocking Smad2 function. Interestingly, mutations in human TGIF gene are associated with holoprosencephaly, suggesting its function in axial midline formation, presumably through regulating Nodal/Smad2 signaling pathway. Based on these results, we will pursue the following specific aims in the proposed research: I) Analysis of the Cripto3-loxP/CriptoNull mice as a model system for ventral forebrain defects and HPE by detailed analysis of the defects in Cripto3-loxP/CriptoNull mice at morphology and molecular levels; II) Investigation of mechanisms underlying Cripto function in mouse axial midline formation by identifying the tissues and cells where Cripto is functioning and downstream target genes of Cripto; III) Investigation of TGIF function in mouse axial midline development by generating TGIF null embryos and examining the modulation of Nodal signaling by TGIF. These studies should improve our understanding of mammalian axial midline formation and human holoprosencephaly.